WO2023148929A1

WO2023148929A1 - Container sorting system

Info

Publication number: WO2023148929A1
Application number: PCT/JP2022/004454
Authority: WO
Inventors: 凜大朗清水; 義人平井; 雅信本江
Original assignee: 株式会社Pfu
Priority date: 2022-02-04
Filing date: 2022-02-04
Publication date: 2023-08-10

Abstract

Provided is a container sorting system, wherein: a camera 20 captures images of a waste group on a conveyor belt conveying the waste group; an image recognition unit 12 recognizes a desired container in the waste group based on the images captured by the camera 20; a first extraction unit 31a extracts the desired container from the waste group conveyed on the conveyor belt; a presence/absence determination unit 15 determines whether any object is present in the extracted desired container; and a sorting device control unit 17 controls whether to have the first extraction unit 31a perform an operation to return the extracted container onto the conveyor belt based on whether any object is present in the extracted container.

Description

Container sorting system

The present disclosure relates to a container sorting system.

At the waste disposal site, a large amount of waste flows on a belt conveyor and is processed every day. At sites where waste is processed, sorting of waste is performed manually. While sorting waste is a simple task, it places a heavy burden on workers who sort waste (hereafter sometimes referred to as "sorters"). A device (sometimes referred to hereinafter as a "waste sorter") has been developed to do so.

JP-A-11-193105 JP 2019-181573 A Japanese Patent Publication No. 2021-523026

When the waste sorting device is made to do the work that the sorting worker was doing instead of the sorting worker, the waste sorting device recognizes each waste flowing on the belt conveyor, and based on the recognition result, the suction pad etc., to extract the desired waste (hereinafter sometimes referred to as "desired waste") from the waste group flowing on the belt conveyor.

Here, if the desired waste is a bottle, there may be leftover liquid or the like in the bottle. A bottle in which an object such as a liquid is present (hereinafter sometimes referred to as a "bottle with an object") is not subject to recycling and therefore excluded from collection.

Therefore, this disclosure proposes a technique that can avoid collection of non-recyclable containers.

A container sorting system of the present disclosure includes a camera, a recognition section, a first extraction section, a first determination section, and a control section. The camera captures an image of the waste group along the transport path along which the waste group is transported. The recognition unit recognizes a desired container included in the waste group based on the image. The first extractor extracts the desired container from the waste group conveyed on the conveying path. The first determination unit determines whether or not an object exists in the extracted container, which is the desired container that has already been extracted. The control unit controls whether or not to cause the first extraction unit to return the extracted container to the conveying path based on whether or not the object exists in the extracted container. .

According to the disclosed technology, it is possible to avoid collecting non-recyclable containers.

FIG. 1 is a diagram illustrating a configuration example of a container sorting system according to Example 1 of the present disclosure. FIG. 2 is a diagram illustrating a configuration example of a control device and a sorting device according to Example 1 of the present disclosure. FIG. 3 is a diagram illustrating a configuration example of an extraction unit according to the first embodiment of the present disclosure; FIG. 4 is a diagram illustrating a configuration example of an extraction unit according to the first embodiment of the present disclosure; FIG. 5 is a diagram illustrating an operation example of an image recognition unit and an area centroid calculation unit according to the first embodiment of the present disclosure; FIG. 6 is a diagram illustrating an operation example of the space determination unit according to the first embodiment of the present disclosure; FIG. 7 is a diagram illustrating an operation example of a space determination unit according to the first embodiment of the present disclosure; FIG. 8 is a diagram illustrating an operation example of the space determination unit according to the first embodiment of the present disclosure; FIG. 9 is a diagram illustrating an example of a presence/absence determination table according to the first embodiment of the present disclosure; FIG. 10 is a diagram illustrating a configuration example of a control device and a sorting device according to a second embodiment of the present disclosure; FIG. 11 is a diagram illustrating an operation example of a presence/absence determination unit according to the second embodiment of the present disclosure; FIG. 12 is a diagram illustrating a configuration example of a container sorting system according to Example 3 of the present disclosure. FIG. 13 is a diagram illustrating a configuration example of a control device and a sorting device according to a third embodiment of the present disclosure;

Hereinafter, embodiments of the present disclosure will be described based on the drawings. The same symbols are attached to the same configurations in the following embodiments.

[Example 1]
<Configuration of container sorting system>
FIG. 1 is a diagram illustrating a configuration example of a container sorting system according to Example 1 of the present disclosure.

1, the container sorting system 1 includes a control device 10, a camera 20, a sorting device 30, a belt conveyor 40, a first collection box 51, a second collection box 52, a first disposal box 53, and a second disposal bin 54 . The sorting device 30 has a first extraction section 31a and a second extraction section 31b. The control device 10, camera 20, and sorting device 30 are connected to each other via a network.

In the following, a case where the container sorting system 1 shown in FIG. That is, hereinafter, the case where the objects to be sorted by the container sorting system 1 are waste will be described as an example. Further, the case where the desired waste is a bottle will be described below as an example. A bottle is an example of a container, and the technology disclosed herein can also be applied to containers other than bottles. In other words, objects to be sorted by the container sorting system 1 are not limited to bottles, and the container sorting system 1 can be used for various containers.

The belt conveyor 40 conveys the waste group placed on the belt conveyor 40 in the conveyance direction CD (+X direction). That is, the belt conveyor 40 forms a transport path along which the waste group is transported in the transport direction CD. The waste group includes bottles and objects other than bottles. The bottles included in the waste group are bottles of desired waste (hereinafter sometimes referred to as "desired bottles") and bottles of undesired waste (hereinafter sometimes referred to as "undesired bottles"). ) and

The camera 20 is arranged above the belt conveyor 40 on which the waste group is conveyed, has a predetermined angle of view, and photographs a predetermined area on the upper surface of the belt conveyor 40 from above the belt conveyor 40 at a constant frame rate. do. Therefore, the image captured by the camera 20 is an image of the waste group (hereinafter sometimes referred to as a "waste group image"). A waste group image is transmitted from the camera 20 to the control device 10 .

The control device 10 controls the operation of the sorting device 30 based on the waste group image, and causes the first extraction unit 31a or the second extraction unit 31b to extract the desired bottle from among the bottles included in the waste group.

Under the control of the control device 10, the sorting device 30 sorts the bottles by extracting them from the waste group conveyed in the conveying direction CD by the belt conveyor 40, and removes the extracted bottles outside the belt conveyor 40. transport. The sorting device 30 extracts bottles from the waste group using the first extraction part 31a and the second extraction part 31b.

The first collection box 51 , the second collection box 52 and the first disposal box 53 are installed along the side of the belt conveyor 40 . The first collection box 51 is installed within the movable range of the first extraction part 31a in the X-axis direction, and the second collection box 52 is installed within the movable range of the second extraction part 31b in the X-axis direction. The first disposal box 53 is installed within the movable range of the first extractor 31a and the second extractor 31b in the X-axis direction. Further, the second disposal box 54 is arranged at the end of the belt conveyor 40 in the X-axis direction, and the first collection box 51, the second collection box 52, and the first disposal The waste remaining on the belt conveyor 40 without being thrown into the box 53 is thrown from the belt conveyor 40 into the second disposal box 54. - 特許庁

<Configuration of control device and sorting device>
FIG. 2 is a diagram illustrating a configuration example of a control device and a sorting device according to Example 1 of the present disclosure. In FIG. 2 , the control device 10 has an image processing section 11 , a presence/absence determination section 15 , a learned model storage section 16 , and a sorting device control section 17 . The image processing unit 11 has an image recognition unit 12 , an area centroid calculation unit 13 , and a space determination unit 14 . The sorting device 30 has a first extraction section 31a and a second extraction section 31b. The first extraction part 31a has a first measurement part 61a, and the second extraction part 31b has a second measurement part 61b. A waste group image captured by the camera 20 is input to the image processing section 11 . The first measurement unit 61a measures the weight of the desired bottle extracted by the first extraction unit 31a, and the second measurement unit 61b measures the weight of the desired bottle extracted by the second extraction unit 31b. Hereinafter, the first extraction section 31a and the second extraction section 31b may be collectively referred to as the "extraction section 31", and the first measurement section 61a and the second measurement section 61b may be collectively referred to as the "measurement section 61".

<Structure of extraction unit>
3 and 4 are diagrams illustrating configuration examples of an extraction unit according to the first embodiment of the present disclosure. For example, the extractor 31 is formed using a suction pad 311, as shown in FIGS. The sorting device 30 sorts out the desired bottles from the waste group conveyed on the belt conveyor 40 using the extraction unit 31 under the control of the sorting device control unit 17 . The extraction unit 31 is movable in the ±X, ±Y, and ±Z directions, and lifts the bottle extracted from the waste group in the +Z direction and moves it in the ±X and ±Y directions. , first collection box 51 , second collection box 52 or first disposal box 53 . The first extraction unit 31a puts the desired bottle extracted from the waste group into either the first collection box 51 or the first disposal box 53, and the second extraction unit 31b extracts it from the waste group. The desired bottle thus obtained is put into either the second collection box 52 or the first disposal box 53 . In addition, the second extraction unit 31b is arranged downstream of the first extraction unit 31a in the conveying direction CD, and after the desired bottles are extracted by the first extraction unit 31a, the waste bottles are conveyed on the belt conveyor 40. Extract other desired bottles from the object group.

Further, as shown in FIG. 3, for example, a strain sensor 611 is installed on the column 312 of the extractor 31 . The larger the weight of the bottle extracted from the waste group by the suction pad 311 (hereinafter sometimes referred to as the “extracted bottle”) (hereinafter sometimes referred to as the “bottle weight”), the greater the weight of the column 312 . Since the amount of extension in the Z-axis direction (hereinafter sometimes referred to as the "support extension amount") increases, the strain sensor 611 measures the weight of the extracted bottle by detecting the extension amount of the support. The strain sensor 611 is an example of the measuring section 61 .

Also, for example,

strain sensors

612a, 612b, 612c, and 612d are installed on the arm 313 of the extraction unit 31, as shown in FIG. Since the deformation amount of the arm 313 (hereinafter sometimes referred to as "arm deformation amount") changes according to the weight of the extracted bottle,

strain sensors

612a, 612b, 612c, and 612d detect the arm deformation amount. Measure the weight of the brewed bottle. The

strain sensors

612 a , 612 b , 612 c and 612 d are examples of the measuring section 61 .

<Operation of Image Recognition Unit and Area Gravity Center Calculation Unit>
FIG. 5 is a diagram illustrating an operation example of an image recognition unit and an area centroid calculation unit according to the first embodiment of the present disclosure;

The image recognition unit 12 recognizes the existence area and type of bottles included in the waste group by image recognition based on the waste group image. Hereinafter, the bottle presence area recognized by the image recognition unit 12 may be referred to as a "bottle area", and the type of bottle recognized by the image recognition unit 12 may be referred to as a "bottle type". The image recognition unit 12 uses the learned model stored in the learned model storage unit 16 to recognize the bottle area and the bottle type. The image recognition unit 12 converts the bottle area and bottle type into the image of the bottle existing in the waste group image (hereinafter sometimes referred to as the "bottle image") as information indicating the characteristics of the bottle (hereinafter referred to as (sometimes referred to as “feature information”).

For example, as shown in FIG. 5, the image recognition unit 12 gives feature information including label information LA indicating the bottle type and contour information CO indicating the bottle area to the bottle image BI. Bottle types are classified into, for example, small energy drink bottles, medium energy drink bottles, large energy drink bottles, wine bottles, champagne bottles, and other bottles. Among the bottles included in the waste group, small energy drink bottles, medium energy drink bottles, large energy drink bottles, wine bottles, and champagne bottles are desired bottles, and the other bottles are undesired bottles. In addition, the contour information CO consists of a plurality of coordinate points (x0, y0), (x1, y1), . formed by That is, an area surrounded by lines connecting a plurality of coordinate points (x0, y0), (x1, y1), .

In addition, the area center of gravity calculation unit 13 calculates the coordinates of the area center of gravity of the bottle (hereinafter sometimes referred to as "area center of gravity coordinates"). The area-center-of-gravity calculator 13 calculates the coordinates of the center of gravity of the bottle region recognized by the image recognition unit 12 as the area-center-of-gravity coordinates. The area-center-of-gravity calculator 13 calculates area-center-of-gravity coordinates AC(Xa, Ya) in the bottle area based on the contour information CO. The area barycentric calculation unit 13 sets the area barycentric coordinates to the coordinates indicating the extraction point when the extraction unit 31 extracts the bottle (hereinafter sometimes referred to as "extraction point coordinates"), and calculates the set extraction point coordinates. is output to the sorting device control unit 17 .

Since the area barycentric coordinates are calculated based on the waste group image, the extraction point coordinates set by the area barycentric calculation unit 13 are based on the coordinate system of the waste group image, that is, the coordinate system of the camera 20 (hereinafter referred to as the “camera coordinates in a coordinate system). Therefore, the sorting device control unit 17 converts the extraction point coordinates in the camera coordinate system set by the area barycenter calculation unit 13 into the extraction point coordinates in the coordinate system of the sorting device 30 . The sorting device control unit 17 outputs a control signal including the extracted point coordinates after conversion to the sorting device 30 . Further, the sorting device control unit 17 causes the second extraction unit 31b to extract bottles that cannot be extracted by the first extraction unit 31a.

The sorting device 30 moves the extraction unit 31 directly above the extraction point coordinates indicated by the control signal, and the extraction unit 31 uses the extraction point coordinates as the extraction point to extract waste from the waste group conveyed on the belt conveyor 40 . Extract the desired bottle from When the extraction part 31 is formed using the suction pad 311 , the extraction part 31 extracts the bottle by suction of the suction pad 311 with the extraction point on the bottle being the suction position of the suction pad 311 with respect to the bottle.

<Operation of the space determination unit>
6, 7, and 8 are diagrams illustrating an operation example of the space determination unit according to the first embodiment of the present disclosure.

The waste group image I1 shown in FIG. 6 includes bottle images BI11, BI12, BI13, BI14, and BI15. A bottle image BI11 is an image of a desired bottle, and bottle images BI12, BI13, BI14, and BI15 are images of an undesired bottle. The space determination unit 14 sets bounding boxes BB11, BB12, BB13, BB14, and BB15 to the bottle images BI11, BI12, BI13, BI14, and BI15, respectively. Further, the space determination unit 14 detects the length of the long side and the length of the short side of the bounding box BB11 set in the bottle image BI11 among the bounding boxes BB11, BB12, BB13, BB14, and BB15, thereby determining the bounding box. The size of box BB11 is detected. Since the bottle B11 that is the subject of the bottle image BI11 is the desired bottle, the bottle B11 is extracted from the waste group by the first extraction unit 31a or the second extraction unit 31b.

A waste group image I2a shown in FIG. 7 is a waste group image taken after the waste group image I1 (FIG. 6) was taken. The waste group image I2a shown in FIG. 7 includes bottle images BI21, BI22, BI23, BI24, and BI25. The space determination unit 14 sets bounding boxes BB21, BB22, BB23, BB24, and BB25 to the bottle images BI21, BI22, BI23, BI24, and BI25, respectively. Further, based on the waste group image I2a, the space determination unit 14 determines a space in which the extracted bottle can be returned between the waste groups on the belt conveyor (hereinafter sometimes referred to as "empty space"). exists. For example, the space determination unit 14 calculates the mutual distances between adjacent sides in the bounding boxes BB21, BB22, BB23, BB24, and BB25 and the waste group image I2a, thereby obtaining a rectangle having no bounding box. , the maximum space (hereinafter sometimes referred to as “maximum space”) is detected. The maximum space in the waste group image I2a is the space MSa. The space determination unit 14 also determines whether or not the bounding box BB11 (FIG. 6) fits within the detected maximum space MSa. When the bounding box BB11 fits within the detected maximum space MSa, the space determination unit 14 determines that there is an empty space for the bottle B11 that is the subject of the bottle image BI11, and determines that the bounding box BB11 exists within the detected maximum space MSa. If the box BB11 does not fit, it is determined that there is no empty space for the bottle B11 that is the subject of the bottle image BI11. Since the bounding box BB11 (FIG. 6) fits within the maximum space MSa shown in FIG. 7, when the waste group image I2a is captured by the camera 20, the space determination unit 14 determines that there is an empty space. .

A waste group image I2b shown in FIG. 8 is another waste group image taken after the waste group image I1 (FIG. 6) was taken. The waste group image I2b shown in FIG. 8 includes bottle images BI31, BI32, BI33, BI34, BI35, and BI36. The space determination unit 14 sets bounding boxes BB31, BB32, BB33, BB34, BB35, and BB36 to the bottle images BI31, BI32, BI33, BI34, BI35, and BI36, respectively. Also, the space determination unit 14 determines whether or not there is an empty space based on the waste group image I2b. For example, the space determination unit 14 detects the maximum space by calculating the mutual distances between adjacent sides in the bounding boxes BB31, BB32, BB33, BB34, BB35, and BB36 and the waste group image I2b. . The maximum space in the waste group image I2b is the space MSb. The space determination unit 14 also determines whether or not the bounding box BB11 (FIG. 6) fits within the detected maximum space MSb. When the bounding box BB11 fits within the detected maximum space MSb, the space determination unit 14 determines that there is an empty space for the bottle B11 that is the subject of the bottle image BI11, and determines that the bounding box BB11 exists within the detected maximum space MSb. If the box BB11 does not fit, it is determined that there is no empty space for the bottle B11 that is the subject of the bottle image BI11. Since the bounding box BB11 (FIG. 6) cannot fit within the maximum space MSb shown in FIG. judge.

<Operation of presence/absence determination unit>
FIG. 9 is a diagram illustrating an example of a presence/absence determination table according to the first embodiment of the present disclosure; The presence/absence determination unit 15 has a presence/absence determination table TA1 shown in FIG. In the presence/absence determination table TA1, associations between bottle types and threshold values THA[g] are set in advance. Based on the bottle type recognized by the image recognition unit 12 and the bottle weight measured by the measurement unit 61, the presence/absence determination unit 15 determines whether the extracted bottle is an object-existing bottle (that is, the extracted bottle (whether or not there is an object in).

For example, the presence/absence determination unit 15 uses the presence/absence determination table TA1 to determine that when the bottle type of the extracted bottle is a "small nutritional drink bottle", the weight of the extracted bottle is equal to or greater than the threshold THA of 74 [g]. Determine that the bottle is an object-existing bottle, and determine that the extracted bottle is not an object-existing bottle (that is, there is no object in the extracted bottle) when the bottle weight is less than the threshold value THA of 74 [g]. . Further, for example, the presence/absence determination unit 15 uses the presence/absence determination table TA1 to extract when the bottle type of the extracted bottle is a "medium-sized nutritional drink bottle" and the weight of the bottle is equal to or greater than the threshold value THA of 102 [g]. It is determined that the finished bottle is an object-existing bottle, and that the extracted bottle is not an object-existing bottle when the bottle weight is less than the threshold value THA of 102 [g]. Further, for example, the presence/absence determination unit 15 uses the presence/absence determination table TA1 to extract when the bottle type of the extracted bottle is a “large nutritional drink bottle” and the weight of the bottle is equal to or greater than the threshold THA of 144 [g]. It is determined that the finished bottle is an object-existing bottle, and that the extracted bottle is not an object-existing bottle when the bottle weight is less than the threshold value THA of 144 [g]. Further, for example, the presence/absence determination unit 15 uses the presence/absence determination table TA1 to determine that when the bottle type of the extracted bottle is "wine bottle", the weight of the extracted bottle is equal to or greater than the threshold value THA of 400 [g]. is determined to be an object-existing bottle, and the extracted bottle is determined not to be an object-existing bottle when the bottle weight is less than the threshold value THA of 400 [g]. Further, for example, the presence/absence determination unit 15 uses the presence/absence determination table TA1 to determine that when the bottle type of the extracted bottle is "champagne bottle", the weight of the extracted bottle is equal to or greater than the threshold value THA of 590 [g]. is determined to be an object-existing bottle, and the extracted bottle is determined not to be an object-existing bottle when the bottle weight is less than the threshold THA of 590 [g].

Here, in setting the threshold THA in the presence/absence determination table TA1, for example, the weight of the empty small-sized energy drink bottle is less than 70 [g], the weight of the empty medium-sized energy drink bottle is less than 98 [g], It was assumed that the weight of an empty large nutritional drink bottle is less than 140 [g], the weight of an empty wine bottle is less than 380 [g], and the weight of an empty champagne bottle is less than 570 [g].

<Operation of Sorting Device Control Unit>
The sorting device control unit 17 controls whether or not to cause the extraction unit 31 to return the extracted bottle to the belt conveyor 40 based on whether the extracted bottle is an object-existing bottle.

For example, when the presence/absence determination unit 15 determines that the extracted bottle is not a bottle with an object (that is, no object exists in the extracted bottle), the sorting device control unit 17 first collects the extracted bottle. The extraction unit 31 is made to perform an operation of throwing into the box 51 or the second collection box 52 . The sorting device control unit 17 causes the first extraction unit 31a to put the extracted bottles into the first collection box 51, and causes the second extraction unit 31b to put the extracted bottles into the second collection box 52. let it happen

On the other hand, when the presence/absence determination unit 15 determines that the extracted bottle is an object-existing bottle (that is, an object exists in the extracted bottle), the sorting device control unit 17 moves the extracted bottle to the belt. The extraction unit 31 is caused to perform the operation of returning to the conveyor 40 .

Here, since the second extraction section 31b is arranged downstream of the first extraction section 31a in the conveying direction CD, the extracted bottle can be placed on the belt conveyor 40 even though there is no empty space on the belt conveyor 40. When the first extraction unit 31a performs the operation of returning to The position of the desired bottle in the image may be shifted. If the position of the desired bottle to be extracted by the second extraction unit 31b deviates from the position of the desired bottle in the waste group image captured by the camera 20, it becomes difficult for the second extraction unit 31b to extract the desired bottle. . Therefore, when the space determination unit 14 determines that there is an empty space, the sorting device control unit 17 causes the first extraction unit 31a to return the extracted bottle to the empty space. The desired bottle returned to the empty space by the first extractor 31 a is conveyed by the belt conveyor 40 and thrown into the second disposal box 54 . On the other hand, when the space determination unit 14 determines that there is no empty space, the sorting device control unit 17 does not cause the first extraction unit 31a to return the extracted bottles onto the belt conveyor 40. The first extraction unit 31a is made to perform the operation of throwing the extracted bottle into the first disposal box 53. The position of the desired bottle to be extracted by the second extraction unit 31b is photographed by the camera 20 by throwing the extracted bottle into the first disposal box 53 by the first extraction unit 31a when there is no empty space. It is possible to prevent the bottle from deviating from the desired position in the waste group image.

In addition, since no extracting unit for extracting the desired bottle is arranged downstream of the second extracting unit 31b in the transport direction CD, the sorting device control unit 17 controls whether or not there is an empty space ( In other words, the second extraction unit 31b is caused to return the extracted bottle onto the belt conveyor 40 (regardless of whether there is an empty space).

The first embodiment has been described above.

[Example 2]
<Configuration of control device and sorting device>
FIG. 10 is a diagram illustrating a configuration example of a control device and a sorting device according to a second embodiment of the present disclosure; In FIG. 10 , the control device 10 has an image processing section 11 , a presence/absence determination section 15 , a learned model storage section 16 , and a sorting device control section 17 . The image processing unit 11 has an image recognition unit 12 , an area centroid calculation unit 13 , a space determination unit 14 , and a cross-sectional area calculation unit 18 . The sorting device 30 has a first extraction section 31a and a second extraction section 31b. The first extraction part 31a has a first measurement part 61a, and the second extraction part 31b has a second measurement part 61b.

In general, the larger the cross-sectional area of the bottle, the greater the weight of the bottle. Therefore, the cross-sectional area calculation unit 18 calculates the cross-sectional area of the desired bottle (hereinafter sometimes referred to as “bottle cross-sectional area”) based on the bottle area recognized by the image recognition unit 12 . For example, the cross-sectional area calculator 18 calculates the cross-sectional area of the bottle by multiplying the area of the bottle region by a predetermined coefficient corresponding to the image size of the waste group image.

<Operation of presence/absence determination unit>
FIG. 11 is a diagram illustrating an operation example of a presence/absence determination unit according to the second embodiment of the present disclosure;

Based on the bottle cross-sectional area calculated by the cross-sectional area calculating unit 18 and the bottle weight measured by the measuring unit 61, the presence/absence determination unit 15 determines whether the extracted bottle is an object-existing bottle (that is, the extracted bottle). (whether or not there is an object in the used bottle).

For example, the presence/absence determination unit 15 has a linear function graph ( FIG. 11 ) showing the correspondence relationship between the bottle cross-sectional area [cm ² ] and the threshold value THB [g]. Calculate THB. For example, when the bottle cross-sectional area is 100 [cm ² ], the presence/absence determining unit 15 calculates the threshold THB as 350 [g].

Then, the presence/absence determining unit 15 determines that the extracted bottle is an object-existing bottle when the bottle weight is equal to or greater than the threshold THB, and determines that the extracted bottle is not an object-existing bottle when the bottle weight is less than the threshold THB. There is no object in the extracted bottle).

Here, in setting the graph of the linear function shown in FIG. 11, for example, it is assumed that the weight of an empty bottle with a bottle cross-sectional area of 100 [cm ² ] is 300 [g].

The second embodiment has been described above.

[Example 3]
<Configuration of container sorting system>
FIG. 12 is a diagram illustrating a configuration example of a container sorting system according to Example 3 of the present disclosure. In the container sorting system 2 of the third embodiment, the first disposal box 53 is not provided, and the belt conveyor 40 has three areas: the first area R1a, the second area R2, and the third area R1b. It differs from the container sorting system 1 of the first embodiment. In FIG. 12, illustration of the control device 10, the camera 20, and the sorting device 30 of the container sorting system 2 is omitted for easy understanding of the explanation.

<Configuration of control device and sorting device>
FIG. 13 is a diagram illustrating a configuration example of a control device and a sorting device according to a third embodiment of the present disclosure; In FIG. 13 , the control device 10 has an image processing section 11 , a presence/absence determination section 15 , a learned model storage section 16 , and a sorting device control section 17 . The image processing unit 11 has an image recognition unit 12 and an area centroid calculation unit 13, but does not have a space determination unit 14 (FIG. 2). The sorting device 30 has a first extraction section 31a and a second extraction section 31b. The first extraction part 31a has a first measurement part 61a, and the second extraction part 31b has a second measurement part 61b.

Here, in the belt conveyor 40, the first region R1a is a predetermined region at one end of the belt conveyor 40 in the Y-axis direction, and the third region R1b is a predetermined region at the other end of the belt conveyor 40 in the Y-axis direction. is a predetermined area at the end of the . Below, the first region R1a and the third region R1b may be collectively referred to as "edge regions". On the other hand, the second region R2 is a region other than the end regions in the Y-axis direction of the belt conveyor 40 . The second area R2 is preset as an area where the waste group can be placed, while the end area is preset as an area where the waste group is prohibited. Thus, there is always free space in the edge region.

Therefore, the sorting device control unit 17 causes the extraction unit 31 to return the extracted bottle to the end region of the belt conveyor 40 when the presence/absence determination unit 15 determines that the extracted bottle is an object-existing bottle. Let

The third embodiment has been described above.

[Example 4]
If the object in the desired bottle is liquid, the liquid shakes when the desired bottle is extracted from the waste group, resulting in a large error in the weight of the bottle measured by the measuring unit 61. There is

Therefore, the measuring unit 61 measures the weight of the desired bottle extracted by the extracting unit 31 multiple times. The measurement unit 61 also determines the average value of bottle weights measured a plurality of times (hereinafter, sometimes referred to as “bottle weight average value”) as the final bottle weight.

The sorting device control unit 17 controls whether or not to cause the extraction unit 31 to return the extracted bottles to the belt conveyor 40 based on the bottle weight average value.

The sorting device control unit 17 instructs the extraction unit 31 to return the extracted bottle to the belt conveyor 40 when the difference between the maximum value and the minimum value of bottle weights measured multiple times is equal to or greater than the threshold THC. Instead, the extraction unit 31 is caused to perform the operation of throwing the extracted bottle into the first disposal box 53 . When this causes a large error in the bottle weight measured by the measurement unit 61, it is difficult for the control device 10 to determine whether or not the extracted bottle is an object-existing bottle. In addition, it is possible for the sorting operator to confirm whether or not the extracted bottle is an object-existing bottle.

The fourth embodiment has been described above.

[Example 5]
If the desired bottle has a cap, there is a high probability that liquid is present in the desired bottle. Therefore, when the image recognition unit 12 detects that the desired bottle has a cap, the measurement unit 61 increases the number of times the bottle weight is measured compared to when the desired bottle does not have a cap. For example, the measuring unit 61 measures the weight of a desired bottle without a cap five times and determines the average value of the weights of the five measurements as the final bottle weight, whereas the weight of the desired bottle with a cap is determined. The weight is measured 10 times and the average value of the 10 weight measurements is determined as the final bottle weight.

The fifth embodiment has been described above.

[Example 6]
The trained model storage unit 16 is implemented as hardware, for example, by memory or storage. The image processing unit 11, the presence/absence determination unit 15, and the sorting device control unit 17 include hardware such as a CPU (Central Processing Unit), DSP (Digital Signal Processor), FPGA (Field Programmable Gate Array), ASIC (Application Specific Integrated Circuit) or the like.

Further, all or part of each processing in the above description in the image processing unit 11, presence/absence determination unit 15, and sorting device control unit 17 may be realized by causing a processor to execute a program corresponding to each processing. For example, a program corresponding to each process described above may be stored in the memory of the control device 10, and the program may be read from the memory and executed by the processor. The program is stored in a program server connected to the control device 10 via an arbitrary network, downloaded from the program server to the control device 10 and executed, or stored in a recording medium readable by the control device 10. and may be read from the recording medium and executed. Recording media readable by the control device 10 include, for example, memory cards, USB memories, SD cards, flexible disks, magneto-optical disks, CD-ROMs, and portable storage media such as DVDs.

The sixth embodiment has been described above.

In addition, in Examples 1 to 6, as an example, the case where the sorting device 30 has two extraction units, the first extraction unit 31a and the second extraction unit 31b, has been described. However, the technology disclosed is also applicable to a container sorting system in which the sorting device 30 has a single extractor. Furthermore, the disclosed technology can also be applied to a container sorting system in which the sorting device 30 has three or more extraction units.

As described above, the container sorting system of the present disclosure (

container sorting systems

1 and 2 of the embodiment) includes a camera (camera 20 of the embodiment), a recognition unit (image recognition unit 12 of the embodiment), and a first extraction It has a section (first extraction section 31a in the embodiment), a first determination section (presence/absence determination section 15 in the embodiment), and a control section (sorting device control section 17 in the embodiment). The camera captures an image of the waste group on the transport path (belt conveyor 40 in the embodiment) along which the waste group is transported. The recognition unit recognizes a desired container included in the waste group based on the image captured by the camera. The first extractor extracts a desired container from the waste group conveyed on the conveying path. The first determination unit determines whether an object exists in the extracted container, which is the desired extracted container. The control unit controls whether or not to cause the first extraction unit to return the extracted container to the conveying path based on whether or not there is an object in the extracted container.

By doing this, it is possible to avoid collecting containers that are not subject to recycling, such as containers that contain objects such as liquids.

For example, the container sorting system of the present disclosure has a measurement unit (first measurement unit 61a in the embodiment) that measures the weight of the extracted container. The recognition unit recognizes the type of desired container based on the image captured by the camera. The first determination unit determines that an object exists in the extracted container when the measured weight is equal to or greater than a threshold value corresponding to the recognized type.

Further, for example, the container sorting system of the present disclosure has a calculation unit (cross-sectional area calculation unit 18 in the embodiment), and the recognition unit recognizes the existence area of the desired container. The calculation unit calculates the cross-sectional area of the desired container based on the recognized presence area. The first determination unit determines that an object exists in the extracted container when the measured weight is equal to or greater than a threshold value corresponding to the calculated cross-sectional area.

Also, for example, the control unit causes the first extraction unit to return the extracted container to the conveying path when it is determined that there is an object in the extracted container.

Also, for example, when it is determined that an object exists in the extracted container, the control unit causes the first extraction unit to return the extracted container to a predetermined area at the end of the conveying path.

Also, for example, the container sorting system of the present disclosure has a second determination unit (space determination unit 14 in the embodiment). The second determination unit can return the extracted container between the waste groups based on the image of the waste group captured by the camera after the desired container is extracted by the first extraction unit. Determine whether there is free space. The control unit causes the first extraction unit to return the extracted container to the empty space when it is determined that the empty space exists.

Also, for example, the container sorting system of the present disclosure has a second extraction unit (second extraction unit 31b in the embodiment). The second extraction part is arranged downstream of the first extraction part in the conveying path, and after the desired container is extracted by the first extraction part, other desired wastes are extracted from the waste group conveyed in the conveying path. extract container. When it is determined that an empty space exists and the second extraction unit is arranged, the control unit causes the first extraction unit to return the extracted container to the empty space.

Further, for example, when it is determined that there is no empty space, the control unit does not cause the first extraction unit to return the extracted container to the conveying path, and places the extracted container in a predetermined box (in the embodiment). The first extractor is made to perform the operation of throwing it into the first disposal box 53).

Further, for example, when the second extraction unit is not arranged downstream of the first extraction unit in the transport path, the control unit causes the extracted container to return to the transport path regardless of whether there is an empty space. Let the one extraction part do it.

Also, for example, the measurement unit measures the weight of the extracted container multiple times. The control unit does not cause the first extraction unit to return the extracted container to the conveying path when the difference between the maximum value and the minimum value of the weight measured a plurality of times is equal to or greater than the threshold value, and into a predetermined box.

1, 2 container sorting system 10 control device 11 image processing unit 12 image recognition unit 13 center of gravity calculation unit 14 space determination unit 15 presence/absence determination unit 16 learned model storage unit 17 sorting device control unit 18 cross-sectional area calculation unit 20 camera 30 sorting Apparatus 31a First extraction unit 31b Second extraction unit 61a First measurement unit 61b Second measurement unit

Claims

a camera that captures an image of the waste group on a transport path along which the waste group is transported;
a recognition unit that recognizes a desired container included in the waste group based on the image;
a first extraction unit for extracting the desired container from the waste group transported along the transport path;
a first determination unit that determines whether or not an object exists in the extracted container that is the desired container that has been extracted;
a control unit for controlling whether or not to cause the first extraction unit to return the extracted container to the conveying path based on whether or not the object exists in the extracted container;
A container sorting system comprising:
a measuring unit that measures the weight of the extracted container,
The recognition unit recognizes the type of the desired container based on the image,
The first determination unit determines that the object exists in the extracted container when the measured weight is equal to or greater than a threshold value corresponding to the recognized type.
The container sorting system according to claim 1.
The recognition unit recognizes the presence area of the desired container,
a calculation unit that calculates the cross-sectional area of the desired container based on the recognized presence area;
a measuring unit that measures the weight of the extracted container;
further comprising
The first determination unit determines that the object exists in the extracted container when the measured weight is equal to or greater than a threshold value corresponding to the cross-sectional area.
The container sorting system according to claim 1.
The control unit causes the first extraction unit to return the extracted container to the conveying path when it is determined that the object exists in the extracted container.
The container sorting system according to claim 1.
The control unit causes the first extraction unit to return the extracted container to a predetermined area at the end of the conveying path when it is determined that the object exists in the extracted container. ,
The container sorting system according to claim 1.
It is possible to return the extracted container between the waste groups based on the image of the waste group taken by the camera after the desired container is extracted by the first extraction unit. further comprising a second determination unit that determines whether or not there is an empty space,
The control unit causes the first extraction unit to return the extracted container to the empty space when it is determined that the empty space exists.
The container sorting system according to claim 1.
Extracting another desired container from the waste group transported on the transport path after it is determined that the empty space exists and the desired container is extracted by the first extraction unit. When the second extraction section is arranged downstream of the first extraction section in the transport path, the control section causes the first extraction section to return the extracted container to the empty space. ,
The container sorting system according to claim 6.
When it is determined that the empty space does not exist, the control unit places the extracted container in a predetermined box without causing the first extraction unit to return the extracted container to the conveying path. cause the first extraction unit to perform an operation of inputting;
The container sorting system according to claim 6.
A second extraction unit for extracting another desired container from the waste group conveyed on the conveying path after the desired container has been extracted by the first extracting unit on the conveying path. If the first extraction unit is not arranged downstream, the control unit causes the first extraction unit to return the extracted container to the conveying path regardless of the presence or absence of the empty space.
The container sorting system according to claim 6.
The measuring unit measures the weight a plurality of times,
without causing the first extraction unit to return the extracted container to the conveying path when a difference between the maximum value and the minimum value of weights measured a plurality of times is equal to or greater than a threshold value; , causing the first extraction unit to perform an operation of putting the extracted container into a predetermined box;
The container sorting system according to claim 2.